Thoughts from the ocean

Written by Christophe Limpalair on 07/31/2015

I spent my 4th of July with waves crashing down against my bare, sizzling feet as I waited for a massive fish to bite the hook. Unfortunately no massive fish was caught but I did end up realizing something far more valuable.

I realized that ScaleYourCode has far more potential than I ever expected and I'm humbled by the feedback the show has been receiving. All thanks to you!

Tweets about ScaleYourCode

Reddit comments about ScaleYourCode

So much has changed the past 6 months. Speaking of the last 6 months, did you realize we're already halfway through 2015? Yeah, I know.

Who am I?
My name is Christophe Limpalair, and I was born in France. I traveled across the Atlantic when I was 11 years old to live in the U.S., and proceeded to spend most of my time on computers. It started out with video games, of course. I was hooked on Starcraft I and quickly joined clans. If you ever played SC1 and chatted on the Internet client, you know exactly what I'm talking about. Some clans were strictly for gaming, others were to mess around, and then there were warring clans.

Starcraft I Screenshot

I got into this warring clan which was full of hackers. I don't necessarily mean hacker in the media sense of the term, though there were a few of those, but more in the curious sense of the term. By curious I mean interested in figuring out how things work and then setting out to create something new. I was inspired by this and it ignited an interest in programming.

Like many others, I started out by reverse engineering HTML and CSS. Then I moved on to Visual Basic because I wanted to create a bot for Starcraft channels. This chat bot let its user customize the way it looked, chat of course, and probably some other useless stuff.

The only problem is that I never got it to connect to the Blizzard servers. So I got bored and moved on to another bot. A warring bot this time.

You're probably confused by what I mean. Let me explain: we had clans, which were created with channels where people could chat. For example you could have a Clan SYC and in that clan you had virtually as many moderators as you wanted (capped by the max number of users per channel).

Starcraft I Channel

Most of the time, these moderators were bots which could be controlled by owners. These bots could kick, ban, and do other admin stuff. If a clan went to war with another clan, you either fought it out by playing a Starcraft match, or you attacked that clan's channel with lots and lots of bots. Hundreds or thousands of bots would flood or load a channel.

Flooding was the act of rapidly entering and exiting a channel while spamming words. As you can imagine, computers would freeze and crash making it a complete pain for that clan to communicate. Morale would get low and the clan would disband after losing members.

Loading was the act of queuing thousands of bots to enter a channel. Since there was a max amount of users permitted in a channel, it was almost impossible to join that channel because as soon as you banned a bot, another entered.

Just like my previous chat bot, this warring bot had all the bells and whistles, but it couldn't connect to servers.

Looking back on those days, I'm willing to bet it wasn't even that hard, I just didn't persevere. I was lazy.

But in any case I moved on to building rigs...

Christophe's PC build

That was fun but expensive.

Fast-forward to today, and things have changed quite a bit. I'm a programmer who understands that when there's a problem there's also a solution.

I feel like there is a problem with my education and ScaleYourCode is my solution.

Scale Your Code
My first college year was in 2011, and I didn't have a clue what I wanted to do. Sure, I knew how to code a few things here and there, but I didn't want to be a coding monkey and I kept reading about all the 20 year old prodigies building multi-million and billion dollar companies thinking I'd never be able to catch up with their skills.

I was completely missing the point.

In Jeff Atwood's interview, he mentions the fact that engineers they hired at StackExchange were fantastic programmers. They had ample technical skills, and could solve the issues at hand. He goes on to say that a lot of people in our industry are that way. So what's the deal? Why aren't we all millionaires?

Jeff answered that question, and there's an article written by Patrick McKenzie back in 2011 that also answers it. The article title is called Don't Call Yourself A Programmer, And Other Career Advice.

You have to create value. It's all about creating value.

Is what you're working on going to increase the company's bottom line? You accomplish this one of two ways: either you create something that decreases cost (automating, for example) or you create something that increases revenue (a new feature, perhaps).

Maybe your schooling experience was different than mine, but I certainly wasn't taught this. I was taught to choose A, B, C, or D, depending on which one answered the question more accurately. I felt like I was missing a massive part of what I needed in order to achieve the goals I had set for myself.

Sure, I can eventually get to those goals with this process but it is overly bloated and I am obsessed with making things faster and more efficient.

ScaleYourCode is my bridge to this gap. It's why I bring on people who know what they're talking about when it comes to performance and reliability. These are skills that get real jobs and provide real value.

Fast-forward to December 2013 when I met a guy through a mutual friend who was working on a social network startup. He was a business man with no technical skills, and needed a developer. At the time, I was working for a real estate company (right?) but looking for more exposure and experience working on real-world apps.

By the way, that real estate job did teach me a lot about people. I use those skills everyday, even when programming. Jeff Atwood talks about the importance of people skills for programmers in his interview.

I don't remember how, but I started getting interested in business after reading quite a few blog posts and books like Getting Real, Rework, and a bunch of others.

Business intrigued me so much because performance is judged by the real market, and not depending on your professor's mood. Although the market certainly has moods of its own, it can make me a lot more money ;)

You either create value or you don't. Creating value is far more interesting to me than memorizing the RC4 Stream Generation, and I'm betting on it being far more valuable for my career.

RC4 Stream Generation diagram

I vividly remember the day I spoke with a man who felt I had ambition, but no clear path to achieve my goals. He looked at me and said "I really believe that you need two passions. Your main passion, and another one that complements it." I thought this was the strangest statement. Aren't really successful founders so passionate about one thing? Well yes and no, in this person's view.

Now I understand. Now I have two passions-- technology and business. I'm insanely passionate about how technology works. Especially how it communicates and how we can optimize services. I'm also very passionate about business, ie: building a product or service that solves a need for people.

The best part is that they both work so well together. Not only am I able to figure out the technical aspects, but I also understand how it ties in with the business side. I don't just code because we want feature XYZ, I code because I care about the consequence it will have on the business long-term.

My interest in business led me to find the Mixergy show. I listened to more hours than I care to count and was in awe at the businesses people had successfully built. If you've never heard of Mixergy, it is by far the best business podcast out there.

I was, however, always a little disappointed when Andrew Warner didn't dig deeper into the technical challenges his tech interviewees faced and overcame. This makes sense since Andrew interviews about the business side of things, but it still left a gap that I could not plug with existing podcasts.

At the time, I was still working at the real estate company and also working on that social networking startup in my spare time after school and work. I liked it because I was learning so much and it was a good mix between business and technology. Unfortunately, that startup wasn't going anywhere and the time spent there became an expense instead of an investment.

In December 2014, I heard a lot about the Laravel framework for PHP and wanted to give it a shot during my Christmas break. I learn best when I can apply something to a project, and I remembered that gap I wanted to fill. So I decided to create a mock version of what a site like that would look like. I want to stress on the word mock, because I didn't really take it seriously at the time.

I started developing it, and absolutely fell in love. It took me just a few days to build a CMS from scratch while also working full time because I couldn't stop. I was a lot more efficient with Laravel, and it just made PHP feel right.

At the same time, I was also learning a lot of other cool stuff to do with Nginx, load balancers, etc.. and I wanted an outlet to write. So I added a blog section to the site and started posting.

At that point, I I was looking at a blank Interviews section with an intense desire to learn more, faster. I started reaching out to people and got my first interview. Everything went uphill from there. Views started coming in, interviewees were really generous with their time and knowledge, and frankly very interested in being interviewed.

I started realizing something really powerful. Not only was I connecting with people I admired, but I was learning a lot from having to sift through content to prepare these interviews and solving the problem I had with my education at the same time. In the process, I could share these findings with everyone in the world interested in listening.

Exponential Growth

How many are interested in listening?
I can tell you how many are currently listening, but I frankly have no idea how many others there are. All I know is that there are plenty more.

While it's impossible to truly count unique visitors (most people accept different IPs as unique), here are some stats since 5 months ago when the show started:

  • Over 650 GB of data transferred from interviews (Just from the site and not YouTube or other platforms, and before Jeff Atwood's interview)

  • Over 30,000 views

In the last month:

  • 200 GB transferred

  • Doubled subscribers

  • Over 8,200 views

Number of views don't account for iTunes and Stitcher subscribers.


It's very easy to handle that amount of traffic, so that's definitely not an issue yet. Page load times are also quite nice thanks to caching etc.. There's always room to improve though and I will be optimizing.

The main issue I see is how much data is being transferred. While this is proof that people really watch the interviews and like the content, I do believe there is room to shave off file sizes. Videos are not always easy to optimize without losing too much quality, but this is an area I'm going to focus on going forward.

What's next?
Considering that I am not the best marketer and that I knew next to no one in the industry until recently, I am happy with these stats. That number of views could (and will) be far better, but it's not terrible for a short amount of time with limited resources.

What's it going to take to reach more people? Well I've got a few plans, but this isn't a marketing blog so I'll spare you the details.

The one thing I would ask, though, is for help. Helping spread the word is really easy-- you can do it with a simple Tweet or re-Tweet. You can also email it or text it to friends and coworkers. It means a hell of a lot to me.

You know what else means a lot to me? The fact that you listen to my podcast and read my blog posts. Thank you, my friend.

On to a bright future for you and me.

- Christophe Limpalair